Background. Virtual reality (VR) simulators for surgical training might possess the properties needed for basic training in laparoscopy. Evidence for training efficacy of VR has been investigated by research of varying quality over the past decade. Objective. To review randomized controlled trials regarding VR training efficacy compared with traditional or no training, with outcome measured as surgical performance in humans or animals. Data sources. In June 2011 Medline, Embase, the Cochrane Central Register of Controlled Trials, Web of Science and Google Scholar were searched using the following medical subject headings (MeSh) terms: Laparoscopy/standards, Computing methodologies, Programmed instruction, Surgical procedures, Operative, and the following free text terms: Virtual real* OR simulat* AND Laparoscop* OR train* Limits: Controlled trials. Study eligibility criteria. All randomized controlled trials investigating the effect of VR training in laparoscopy, with outcome measured as surgical performance. Methods. A total of 98 studies were screened, 26 selected and 12 included, with a total of 241 participants. Results. Operation time was reduced by 17-50% by VR training, depending on simulator type and training principles. Proficiency-based training appeared superior to training based on fixed time or fixed numbers of repetition. Simulators offering training for complete operative procedures came out as more efficient than simulators offering only basic skills training. Conclusions. Skills in laparoscopic surgery can be increased by proficiency-based procedural VR simulator training. There is substantial evidence (grade IA - IIB) to support the use of VR simulators in laparoscopic training.